Over the last decade or so, most of us have slowly found ourselves increasingly dependent on a single type of battery: a lithium battery. This has led to some interesting questions on why lithium has become the preferred choice, and what comes after lithium? In this blog post I attempt to explain some of these issues.
The answer is technical. But I have to make it accessible. This means the simplification of some technical concepts. So bear with me.
The Search For A High-Energy Battery
Imagine that you are a chemist hard at work at trying to find a new battery. Imagine it’s circa 1960 and you are celebrating the anniversary of the lead-acid battery. You like the lead-acid battery, but are convinced that the world needs an extremely high-energy battery.
You’re not sure exactly why you need more energy , but you’re obsessed with the Zeroth law of batteries which states that “The performance of any battery will fall (just) short of our expectations irrespective of the complexity of the device it is powering”. Moreover your 5-year-old neighbor in Cupertino has convinced you that when he grows up he is going to make a computer that you can carry in your backpack, but that it will require a battery with a lot of energy. You know that if you want a lot of energy you need a high-voltage battery. Much higher than the 2 V of the lead-acid chemistry.
So, you do what any decent chemist would do, and decide to drive across the bay to the library at the U. of California campus at Berkeley. You know that the library has a book that has extensive tables that tabulate the potential of various elements and you are sure you can find something that has a high voltage.
After some rummaging, you decide that the best battery in the world has to be one with a lithium anode and a fluorine cathode. The voltage is almost 6 V. You are excited, but also concerned that water breaks down at 1.2 V. You are also pretty sure that water and lithium should not be mixed. You still remember the mini fireworks that happened when you tried that experiment.
You are also sure that you don’t want to mess with fluorine. Your amputated finger on your right arm is evidence of your past attempts at working with fluorine and you have a new appreciation for the miracle of the opposable thumb. You decide to deal with the cathode problem later.
First things first. You need to get lithium to work.
You then find some interesting electrodeposition studies in a PhD thesis by Harris who was guided by Charles Tobias. The thesis has studies of the deposition of various alkali metals (sodium, lithium) in non-water-based solvents. One of these solvents, propylene carbonate, looks promising. You begin to think that maybe you have the ideal anode.
Time To Risk the Remaining Opposable Thumb?
The Modern Li-ion Battery
An easy way to design a battery is to go to the table of standard potentials, pick something that wants to oxidize, and pair it with something that wants to reduce. Lithium is pretty much at one end of that series, making it an ideal anode.
Turns out our hypothetical chemist was being a bit optimistic. The plating of lithium does get easier when using a non-water based solvent, but even after many decades of research there is still no rechargeable lithium-metal based batteries. We have primary batteries with lithium metal (your watch battery, for example), but no commercially available rechargeable one (yet).
The problem occurs when you charge the battery (which involves plating lithium). Lithium, unlike many other metals, does not plate uniformly, but in the form of sharp needles or dendrites. Dendritic lithium can react and can go through the separator and short with the cathode. This makes the battery go Boom!
Fast forward from 1960 to the late 1980’s and it was becoming obvious (I assume, because I was not there) that the problem of dendrites appeared to be a showstopper. In the mean time, it was clear that instead of plating lithium, one could insert the lithium into a cage, with graphite being a very good cage.
The lithium could reversibly move in and out of the cage. The cage occupies volume and it has weight to it, but it did not (under normal conditions) form dendrites. The voltage at which the lithium moved in and out of graphite was similar to that for lithium.
Our hypothetical chemist had finally found the anode, except it was not on the tables in the library.
Something similar could be done on the cathode side, with John Goodenough from the U. of Texas showing that materials like cobalt oxide were able to insert lithium. The voltage for cobalt oxide was nowhere close to that for the fluorine system, but there were (and there still are) no solvents that can allow operation at the voltages where fluorine works.
Like the graphite anode, cobalt oxide allows us to make a workable cathode. Cobalt oxide, like graphite, is a cage that has weight and volume, but working with it is much more practical than working with lithium and fluorine.
Thus was born the modern lithium-ion battery and the revolution starting with Sony’s commercialization of this technology. The voltage is much lower than our hypothetical chemist’s dream (3.7 V instead of 6 V). And it has these two cages that make the weight and volume of the battery go up. Both of these mean that the battery we use today is considerably worse than the dream battery.
Future of Battery Research
We have to either find a way to increase the voltage, or find a way to increase the capacity.
In lithium ion batteries this means that we are constantly searching for anodes and cathodes that hold more lithium for less weight and volume of the cage. We are also looking for electrolytes that allow us to get closer and closer to the voltage of the dream battery.
So where are we compared to where we can be?
Today we use cathodes that operate at ~3.8 V with a capacity of ~180 mAh/g (I’m simplifying things a bit here). Without getting into details, we could conceivably increase the voltage by 10 to 15 percent and the capacity by 50 percent.
The choices on the anode side are a bit more limiting. We have one material that has a lot more capacity than graphite (10 times more), but it decreases the voltage of the battery.
And just so there is no confusion, increasing the capacity of the anode by 10 times does not increase the energy of the battery that much. I make this point in my blog post “In batteries 2+2=1. Actually more like 1/2. Well… Maybe a little bit less“.
So what option do we have after that? For one, we can remove the cages (which gets us back to what our hypothetical chemist proposed). Further, we don’t need to stick to lithium.
Remember that voltage and capacity are what matter. If we can manage to find something else that has either a higher voltage and/or a higher capacity, then we are onto a new level of evolution of batteries.
Lithium batteries are by no means done evolving. It amuses me that we are already getting impatient about wanting to develop something else. But I can see the merit in starting to think about the future, but to paraphrase Robert Frost, we have “miles to go before we sleep”.
When our hypothetical chemist decided we needed a better battery and went looking for lithium, he was onto something. Something that would revolutionize the consumer-electronics space. Turns out we still need a better battery. And it may still involve going to the library to look at a table.
Venkat Srinivasan is a Staff Scientist at Lawrence Berkeley National Lab and writes about batteries on his site This Week In Batteries.
Image courtesy of Argonne National Laboratory,